Main Menu

Searching for the last unicorn

It was one of the most startling animal discoveries of recent years but in a tragically short space of time the Saola is already one of earth’s rarest animals.

Characterised by its dark coat and striking white facial patterns the Saola (sow-la) has long straight horns, which, when viewed from the side can resemble a single horn – leading to its nickname as the ‘Asian unicorn’.

Saola were only discovered in 1992 when a joint biodiversity survey carried out by WWF with the Vietnamese Ministry of Forestry uncovered some intriguing remains in north-central Vietnam. Later that year the distinctive skull they found with its long straight horns was declared a new species by WWF - Pseudoryx nghetinhensis.

But in just over 20 years since its discovery the Asian unicorn is already under threat of extinction from hunting and the widespread destruction of its habitat; estimates of how many of these rare and gentle creatures remain range from a few hundred to as low as a few dozen.

Rarely sighted at all, the Saola are to be found exclusively in the broadleaf rainforests of the Annamite mountain range, which stretches 1,100 km between Laos and Vietnam.

But just finding their whereabouts is a huge challenge – in fact the animal’s situation is so dire that the next Saola found may well be taken into captivity to attempt to establish a breeding population.


 “Tracking down endangered species is a story of frustration”  explains ecologist Professor Douglas Yu, “partly because of the inaccessible places in which they live”.

Standard ecological methods require skilled field biologists, making multiple visits to a site sometimes over lengthy periods of time to observe the animals or to find their sign.

“Getting to remote places is difficult and dangerous, and needing to carry out systematic surveys over time makes the challenge nearly impossible“ - something which drove Yu to question that surely technology could ease some of the burden?

As a result, UEA scientist Yu is working on a surprising new way of finding evidence of these elusive animals.

“I’ve been wanting to do something like this my whole career, and I’ve finally found a way to do it”

Working in collaboration with the Saola Working Group and Professor Thomas Gilbert at the University of Copenhagen, Douglas Yu and a team of scientists are instead using a technology called metabarcoding to look for the critically endangered animal in the leeches that feed on it. Leeches provide an ideal model for this as they are easy to collect in bulk in the right seasons.


Metabarcoding describes a way of speeding up such ecological investigations by using technology to sift through all of the DNA present in a sample from any given site, in order to get raw data on the presence of species.

Metabarcoding: how it works

  • Samples are collected and processed into ‘DNA soup’
  • Scientist extracts and reads the DNA
  • The DNA sequences from the sample are matched to online DNA reference databases, to allow identification of species
  • Large numbers of specimens can be sequenced together in a single reaction
  • The process is quicker, cheaper, and more accurate than traditional census methods

The Metabarcoding pipeline:

The metabarcoding pipeline

Image reproduced with permission from NatureMetrics

In Laos and Vietnam, researchers have so far collected over 15,000 leeches from the Saola’s native habitat, which are then blended into a thick sludge from which DNA is extracted, amplified, sequenced, and compared against online databases that contain DNA sequences from known mammal species.                                                                               

“The amazing thing is that a leech retains a tiny bit of the DNA from the last animal it fed on, even if that last feed was weeks ago."

"As a result, a single day’s leech collection is a bit like walking up and down that trail day and night for several weeks. Current DNA sequencing technology makes those fragments of DNA available to us now,” Yu comments.

Moreover, because leeches feed on a wide range of species, it is possible to census mammals, birds, frogs, and reptiles from the same collection. In contrast, to survey that wide a range of species using traditional methods would require multiple field biologists.

However, it’s fair to say this new process has been a learning curve. Once the DNA is sequenced, you need a confirmed DNA sequence in the database to match to: the solution lies in building robust DNA databases, but these are still incomplete, making it difficult to correctly identify some of the sequences that are isolated from leeches.

“Imagine having a document in a language you’ve never seen before,” laughs Yu “and the only dictionary you have is incomplete, with misspellings and multiple meanings! You have a DNA sequence, but what is it really?”

Indeed results have sometimes raised eyebrows: Yu once found DNA sequences that matched to arctic bearded seals, from a DNA sample which came from South East Asia. 

“Some of our sequences give us identifications that we know can’t be true, and that is very useful, as it helps us to identify subtle errors in the database or in our own analytical methods.”

These incorrect identifications are due in part to incomplete DNA reference databases. Seals are carnivores, so the supposed bearded-seal DNA is almost certainly a carnivore species that is not currently in the database.  Southeast Asian rainforests are full of small predators, from civets to small cats to linsangs to mongooses. “I certainly wouldn’t want to be a rodent in such a forest.” 

To get around the problem of incomplete reference databases, Yu and colleagues are developing ‘analytical pipelines,’ a painstaking process of several years to build a suite of software tools to give rapid and robust results. Yu believes the research is at a point where scientists can begin to reap the rewards of their endeavours - the code and databases will all eventually be published in the public domain as open science.


It’s not just searching for endangered species that Yu is interested in however - using metabarcoding potentially improves environmental governance in general.  Another project, utilising the same techniques, is concerned with making sense of vast and changing populations of wildlife across huge tracts of conservation areas.

In Ailaoshan National Nature Reserve, a long chain of mountains in Yunnan, China, Yu has helped equip over 100 forest rangers with leech collection kits to use throughout the reserve, which is the size of Singapore.

In collaboration with the Forestry Department, Yu is using metabarcoding to make it possible to carry out annual bio audits of the wildlife there, helping to produce a performance indicator for the reserve as a whole.

Over a month, rangers collected leech samples during their normal patrols, therefore covering the entire reserve. Each ranger was given a kit with an ID card and collection tubes containing a preservative solution in which to pop the leeches. The most productive ranger collected 380 in one walk but most averaged 30-40 each. The DNA analysed so far has detected Asiatic black bear, leopard cats, muntjak deer, wild boars and many rodents and frogs to name but a few, and the Forestry Department returned in 2016 to continue the collection.

Each ranger’s leech collection gives us detection events, or not, for multiple species at once. By mapping each species' detections, a species’ range is estimated, and over years, a species' range expansion or contraction gives an indicator of its population viability.

Infographic: (click to expand):


While the team are far from being able to say if Saola DNA has been recorded, Yu and his colleagues feel they are getting closer to this enigmatic animal.

But what is clear is that technology is changing the way nature is managed, potentially enabling a broader section of the population to carry out ecological investigations. Metabarcoding enables scientists to generate data across hundreds or even thousands of species in just a few months, giving us a wealth of information.



Prof Douglas Yu

My research looks at cooperation between humans and nature (Conservation) and cooperation between species (Mutualisms).

View profile